The Effect of Temperature and Flow Rate on Aluminum Etch Rates in RF Plasmas

A two~dimensional transport and reaction model of a high pressure (-1 torr) high frequency (13.56 MHz) single-wafer parallel plate plasma reactor was developed. The chemical etching uniformity was studied as a function of reactor operating conditions. The ratio of the reactivity of the surrounding electrode surface as compared to that of the wafer surface, S, critically affected uniformity. A bullseye clearing pattern was predicted for S < 1, and the reverse pattern for S > 1, while etching was uniform for S = 1. In the case of S ~ 1, and for the parameter range studied, the absolute uniformity was found to improve by surrounding the wafer with a material of similar reactivity, by increasing the flow rate, or by decreasing the reactor pressure or power. However, such actions also served to decrease the etch rate. The oxygen plasma was used as a model experimental system to test the theoretical predictions. An experimental technique based on spatially resolved optical emission spectroscopy in concert with actinometry and the Abel transform was developed to obtain a three-dimensional mapping of the reactant (O atom) concentration profile in the plasma reactor. When a reactive film was covering part of the substrate electrode, a profound decrease in the reactant concentration was observed over the film. At the same time, large concentration gradients developed, especially at the boundary of the reactive film with the surrounding electrode. Good quantitative agreement was found between the model predictions and the experimental reactant concentration data for the range of pressure, power, flow rate, and reactive film radius examined.

Section: Atandt Bell Laboratories Assisted In Meeting the Publication Cmentioning

confidence: 99%

“…Oz Equation [18] is a symmetry condition and Eq. [19] is a simplified exit boundary condition. Equation [20] accounts for both etchant convection away from and etchant recombination on the upper electrode surface.…”

Section: Ormentioning

confidence: 99%

Uniformity of Etching in Parallel Plate Plasma Reactors

Economou

Park

Williams

1989

“…Etch rate measurements were complicated by the fact that film etching was not uniform. Films etched faster at the edges so that clearing proceeded from the outside edge to the center, consistent with aluminum etching studies (20). The etch time was defined as the time when the entire sample cleared.…”

Section: Resultsmentioning

confidence: 61%

“…To minimize such effects, silver paste was used to thermally bond samples to the electrode. In this way, heat transfer from the sample to the electrode is improved, and the sample temperature should remain near the electrode temperature (20). Figure 4 shows the effect of varying the lower electrode temperature on the etch rate for both bonded and unhonded samples in the flow reactor.…”

Section: Resultsmentioning

confidence: 99%

Plasma‐Enhanced Etching of Tungsten and Tungsten Silicide in Chlorine‐Containing Discharges

Fischl

Hess

1987

Results are presented for the etching of W and WSix in Cl2 and Cl2/BCl3 plasmas. Etch rates were measured for conditions ranging from 0.1 to 1.0 torr pressure, 30° to 170°C electrode temperature, 0.2 to 1.0 W/cm2 power density, and 3 to 200 sccm flow rate. Etch rates varied from below 10 nm/min to 90 nm/min for tungsten and to 450 nm/min for the silicide. Residence time effects on tungsten etching were also studied. Small additions of BCl3 significantly increased the etch rates of both W and WSix and improved the reproducibility. Results suggest that chemical processes have a significant effect on etch characteristics, despite the low vapor pressure of WCl5 and WCl6 .

“…A flowing afterglow reactor has been used to investigate the relative importance of electrode temperature, reactant flow rate, and C1/C12 ratio on aluminum etch rate anduniformity with CC14, BCI3, C12, and mixtures of CC14 and BC13 with C12 (8,32). Comparison of etch results using bonded (epoxied) and unbonded aluminum samples suggests that sample heating due to exothermic chemical reactions and plasma heating may result in a temperature difference of more than 100~ between the unbonded sample and the electrode (32). Results with bonded samples suggest that the sample temperature must drop below 25~ before product volatility is a severe limitation.…”

Section: Steps In the Etching Processmentioning

confidence: 99%

Plasma Etching of Aluminum: A Comparison of Chlorinated Etchants

Danner

Dalvie

Hess

1987

The plasma‐assisted etching of aluminum in chlorine containing RF glow discharges has been studied. Use of a single parallel plate reactor permitted a direct comparison of etch results between BCl3 , BCl3/Cl2 , CCl4 , and SiCl4 . Separation of aluminum etching into native oxide reduction and water vapor/oxygen scavenging, and metal film etching allowed the likely rate‐limiting processes in the etch cycle to be ascertained for the different etch gases. The longer initiation period observed with CCl4 and SiCl4 compared to BCl3 appeared to be due to etch gas dissociation effects. Metal etching was believed to be limited by the removal of CClx and SiClx residues with CCl4 and SiCl4 and by etchant generation with BCl3 .